DOI: 10.1111/jdv.16946 JEADV

GUIDELINES 2020 European guideline on the management of syphilis

M. Janier,1,* M. Unemo,2 N. Dupin,3 G.S. Tiplica,4 M. Potocnik, 5 R. Patel6 1STD Clinic, Hopital^ Saint-Louis AP-HP and Hopital^ Saint-Joseph, Paris, France 2WHO Collaborating Centre for Gonorrhoea and other Sexually Transmitted Infections, Department of Laboratory Medicine, Microbiology, Orebro€ University Hospital and Orebro€ University, Orebro,€ Sweden 3Syphilis National Reference Center, Hopital^ Tarnier-Cochin, AP-HP, Paris, France 42nd Dermatological Clinic, Carol Davila University, Colentina Clinical Hospital, Bucharest, Romania 5Department of Dermatovenereology, University Medical Centre Ljubljana, Ljubljana, Slovenia 6Department of Genitourinary Medicine, the Royal South Hants Hospital, Southampton, UK *Correspondence to: M. Janier. E-mail: [email protected]

Abstract The 2020 edition of the European guideline on the management of syphilis is an update of the 2014 edition. Main modifications and updates include: - The ongoing epidemics of early syphilis in Europe, particularly in men who have sex with men (MSM) - The development of dual treponemal and non-treponemal point-of-care (POC) tests - The progress in non-treponemal test (NTT) automatization - The regular episodic shortage of benzathine penicillin G (BPG) in some European countries - The exclusion of azithromycin as an alternative treatment at any stage of syphilis - The pre-exposure or immediate post-exposure prophylaxis with doxycycline in populations at high risk of acquiring syphilis. Received: 12 June 2020; Accepted: 4 September 2020

Conflicts of interest The authors have no conflicts of interest related to this guideline.

Funding sources None.

Introduction EEA countries and particularly among men who have sex with Syphilis is a systemic human disease due to Treponema pallidum men (MSM).3 subsp. pallidum (referred as T. pallidum below) and classified as This guideline is an update of the 2014 European guideline on acquired or congenital. Acquired syphilis (primarily by sexual the management of syphilis.4 contact) is divided into early and late syphilis. Early syphilis includes primary, secondary and early latent syphilis. The Euro- Case finding (2, C) pean Centre for Disease Prevention and Control (ECDC) defines Routine tests for syphilis should be taken in all pregnant women, early syphilis (infectious syphilis) as syphilis acquired <1 year people donating blood, blood products or solid organs and the previously and the World Health Organization (WHO) as syphi- following groups at higher risk of syphilis: all patients who are lis acquired <2 years previously.1,2 Late syphilis includes late newly diagnosed with sexually transmitted infection (STI); per- latent and tertiary syphilis (gummatous, late cardiovascular and sons with HIV; persons on pre-exposure prophylaxis (PrEP); late neurosyphilis). The ECDC defines late syphilis as syphilis patients with hepatitis B and/or hepatitis C; patients suspected acquired ≥1 year previously and the WHO as syphilis acquired of early neurosyphilis (i.e. unexplained sudden visual loss, unex- ≥2 years previously.1,2 Congenital syphilis (mother-to-child- plained sudden deafness or meningitis); patients who engage in transmission of syphilis) is divided into early (first 2 years) and sexual behaviour that places them at higher risk (e.g. MSM, sex late, including stigmata of congenital syphilis. The incidence of workers and all those individuals at higher risk of acquiring syphilis in the European Union/European Economic area (EU/ STIs). Screening tests should also be offered to all attendees at EEA) has shown an overall increase since 2000, which has been dermato-venereology/genitourinary medicine (GUM)/STI clin- mainly due to a significant increase in Western and Central EU/ ics referred to hereafter as ‘sexual health clinics’.

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 2 Janier et al.

Table 1 Treatment of syphilis in adults Secondary syphilis develops in approximately one-third of Early syphilis (Primary, Secondary and Early latent, i.e. acquired untreated patients, tertiary syphilis in about 10%. Patients are <1 year previously) infectious primarily through sexual contact, mainly in the first First-line therapy option: year (primary and secondary syphilis). Later transmission usu- • Benzathine penicillin G (BPG) 2.4 million units intramuscularly (IM) (one ally by other means (vertically and through tissue donation) is injection of 2.4 million units or 1.2 million units in each buttock) on day 1 well described.10 Second-line therapy option: Incubation period: 10–90 days between infection and emer- • Procaine penicillin 600 000 units IM daily for 10–14 days, i.e. if BPG is not available gence of chancre. Bleeding disorders: Primary syphilis: a chancre, usually with regional lym- • Ceftriaxone 1g intravenously (IV) daily for 10 days phadenopathy. The chancre is typically superficial, single, pain- • Doxycycline 200 mg daily (either 100 mg twice daily or as a single less and indurated with a clean base discharging clear serum, 200 mg dose) orally for 14 days most often in the anogenital region. It is never blistering in Penicillin allergy or parenteral treatment refused: • Doxycycline 200 mg daily (either 100 mg twice daily or as a single appearance. Chancres are often atypical in appearance and may 200 mg dose) orally for 14 days be multiple, painful, deep and indistinguishable from herpetic – Late latent (i.e. acquired ≥1 year previously or of unknown duration), ulceration.11 13 Any anogenital ulcer should be considered cardiovascular and gummatous syphilis syphilitic unless proven otherwise. Chancres are frequently diffi- First-line therapy option: cult to find in females and MSM. Initial tests may not allow a • BPG 2.4 million units IM (one injection 2.4 million units single dose or firm and conclusive rejection of a syphilis diagnosis and retesting 1.2 million units in each buttock) weekly on day 1, 8 and 15 Second-line therapy option: with serology at 1, 2 and 6 weeks is needed to exclude the diag- • Procaine penicillin 600 000 units IM daily during 17–21 days, i.e. if nosis – however, delaying treatment is hazardous in some popu- BPG is not available lations especially when patients are unlikely to return for follow- Penicillin allergy or parenteral treatment refused: up and thorough investigations. • – Some specialists recommend penicillin desensitization as the evidence Secondary syphilis14 19: multisystem involvement due to bacte- base for the use of non-penicillin regimens is weak. • Doxycycline 200 mg daily (either 100 mg twice daily or as a single riaemia, usually within the first year but it may recur in the second 200 mg dose) orally during 21–28 days. year after infection. Usually, non-itching skin rash (roseola in the Neurosyphilis, ocular and auricular syphilis 2–3 months after onset of chancre and papular syphilids later on) First-line therapy option: and/or mucocutaneous lesions are present in 90% of cases. Fever, • Benzyl penicillin 18–24 million units IV daily, as 3-4 million units every generalized lymphadenopathy, hepatitis, splenomegaly, periostitis, – 4 hours for 10 14 day arthritis, aortitis and glomerulonephritis are possible. Meningitis, Second-line therapy option: cranial nerve palsies, auricular and ophthalmic abnormalities (such • If hospitalization and IV benzyl penicillin is impossible • Ceftriaxone 1–2 g IV daily for 10–14 days as uveitis, retinitis, otitis and papillar oedema), meningo-vascular • Procaine penicillin 1.2–2.4 million units IM daily AND probenecid syphilis (stroke, myelitis) can occur in secondary syphilis and 500 mg four times daily, both for 10–14 days should be individualized as early neurosyphilis. Penicillin allergy: Latent syphilis: positive serological tests for syphilis with no • Desensitization to penicillin followed by the first-line regimen clinical evidence of treponemal infection. Rather arbitrarily classi- Pregnancy fied as early if within the first year of infection and late (or unde- First-line option for treatment of early syphilis (i.e. acquired <1 year ≥ previously): termined duration) after 1 year. Early latent syphilis (or non- 6 • BPG 2.4 million units IM single dose (or 1.2 million units in each but- primary non-secondary early syphilis) is a descriptive term that tock) includes patients with positive serological tests for syphilis and: Second-line therapy option: -a negative syphilis serology within 1 year of a syphilis diag- • – Procaine penicillin 600 000 units IM daily for 10 14 days, i.e. if BPG is nosis OR not available Penicillin allergy. -a fourfold (2 dilutions) or greater increase of non-trepone- • Desensitization to penicillin followed by the first-line regimen mal antibodies titre within 1 year of previous testing OR HIV-infected patients -unequivocal evidence that the disease was acquired in the past 20 Treatment of syphilis in patients with concomitant HIV infection year (on the basis of clinical signs in patient and partners). • Treatment should be given as for non-HIV-infected patients. Misclassification of early vs late latent syphilis is common. Tertiary syphilis: Diagnosis -Gummatous syphilis: nodules/plaques or ulcers (skin, mucosae, visceral); 5–9 Clinical -Late neurosyphilis encompasses meningitis, cranial nerve dys- The definition of stages is clinical, chronology beginning with function, meningo-vascular syphilis (stroke, myelitis) and the onset of a chancre (ulcer or erosion). Stages can overlap. parenchymatous neurosyphilis (general paresis, tabes dorsalis);

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 2020 European Syphilis guideline 3

-Cardiovascular syphilis: aortic regurgitation, stenosis of and antigenically similar, and can be differentiated only by their coronary ostia, aortic aneurysm (mainly thoracic). mode of transmission, epidemiology, clinical manifestations and Neurologic syphilis: meningitis, cranial nerve dysfunction, more recently based on minor differences by genomic DNA – can occur early (secondary syphilis) or late (tertiary syphilis) in sequencing.51 53 A person with positive STS should be investi- the course of the disease. gated and treated as for syphilis as a precautionary measure unless previous adequate treatment for syphilis is documented Laboratory (1, D). • Non-treponemal tests (NTT): using a complex antigen con- Demonstration of T. pallidum sisting of cardiolipin, lecithin and cholesterol (lipoidal tests, • Direct detection methods provide definitive diagnosis of reagin tests) such as the Venereal Disease Research Labora- syphilis. tory test (VDRL), the Rapid Plasma Reagin test (RPR) and • Darkfield examination (DFE) of chancres and erosive cuta- the Toluidine Red Unheated Serum Test (TRUST). All these neous lesions was the old gold standard method for defini- tests detect a mixture of heterophile IgG and IgM and are tive diagnosis. It gives immediate results. However, the manually performed, but they are cheap, simple and, if per- method is labour-intensive and subjective and can result in formed appropriately, have a relatively high sensitivity. some false-positive and (many) false-negative results.21,22 NTT usually become positive approximately 10–15 days Due to the availability of more sensitive and specific tests after the appearance of the primary chancre (i.e. around (specifically the PCR), it is not recommended for routine 6 weeks after infection). In the absence of treatment, the diagnosis anymore. titre reaches a peak between 1 and 2 years following infec- • Polymerase chain reaction (PCR) testing is the preferred tion and remains positive with low titres in very late dis- method particularly but not exclusively for oral and other ease.22 Spontaneous seroreversion of NTT in patients with lesions where contamination with commensal treponemes tertiary syphilis is hardly ever observed. Titres of NTT cor- is likely. It can be performed using tissues, cerebrospinal relate grossly with disease activity and are used to monitor – fluid (CSF) or blood (although insensitive in the latter).22 both disease activity and efficacy of treatment. Semi-autom- 28 There is no internationally approved PCR assay for atized RPR tests have been developed.54 However, these T. pallidum and accordingly, it is crucial to select a strictly tests require further optimizations and subsequent evalua- validated and quality-assured method and always use it with tions.55 appropriate quality controls. • Treponemal tests (TT): T. pallidum haemagglutination test • Immunohistochemistry using a polyclonal antibody against (TPHA), T. pallidum particle agglutination test (TPPA), T. pallidum can be efficient to identify treponemes in skin, fluorescent treponemal antibody absorption test (FTA-abs mucosal and tissue lesions,27,28 but it is not suitable for rou- test), treponemal enzyme immunoassay (EIA) or enzyme- tine diagnosis. linked immunosorbent assay (ELISA), chemiluminescence • Hybridization in tissues29 is not used for routine diagnosis. immunoassay (CLIA), IgG or IgM immunoblot test for • Warthin–Starry (argentic) staining on tissues is very diffi- T. pallidum. Most of these tests use recombinant trepone- cult to perform and of limited value in most cases. mal antigens and detect both IgG and IgM. FTA-abs test is • As specialized supplies, equipment and training are becoming obsolete because it is time-consuming, expensive required, the direct fluorescent antibody test is not recom- and difficult to read. TPHA and TPPA are manual and sub- mended for routine diagnosis. ject to variation between individuals in interpretation but • For molecular epidemiological typing, PCR, PCR-restriction they are cheap and widely used all over Europe. EIA/ELISA fragment length polymorphism (RFLP) and/or DNA sequenc- and CLIA tests are most frequently automated but many of ing (e.g. multilocus sequence typing (MLST) or whole genome those remain expensive, suboptimally evaluated and/or sequencing) can be performed on clinical specimens. How- standardized, and some may have suboptimal speci- ever, due to the highly conserved genome of T. pallidum the ficity.22,56 TT mostly become positive in approximately 5– – discriminatory ability of typing methods is in general low.30 36 15 days after emergence of the chancre. Quantitation of TT is not useful in the diagnosis or management of syphilis Serological tests for syphilis (STS)22,37–50 Serological tests for (with possible exception of congenital syphilis). TT should syphilis provide— a presumptive (but essential) diagnosis of therefore not be used to assess disease activity and treatment syphilis. outcome and will remain positive for life in most patients.22 None of the STS differentiate between venereal syphilis and • Specific anti-T. pallidum IgM antibody tests: IgM-EIA, 19S- the non-venereal treponematoses (yaws: T. pallidum subsp per- IgM-FTA-abs test and IgM immunoblot for T. pallidum. tenue; bejel – endemic syphilis: T. pallidum subsp endemicum The sensitivity of such tests is low in active syphilis. IgM and pinta: T. carateum). These pathogens are morphologically does not help to stage syphilis accurately and should not be

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 4 Janier et al.

relied upon to determine the length of treatment. The use- quantitative NTT (reaching at least 1:8 to 1:16 dilution) fulness of measuring IgM in the assessment of newborns should be performed in all cases on the same serum (1, B). and in neurosyphilis has not been appropriately validated.22 Although a confirmatory TT may be important for coun- • Many rapid point-of-care tests (POCTs) using treponemal anti- selling, notification and may have a psychological impact, it gens have been developed in the last 25 years. Initially, tests had has limited impact on treatment.69 In patients with a positive suboptimal sensitivity compared to traditional methods, but TT, a negative NTT and no suspicion of very early syphilis (no some of the latest assays have shown a substantial improve- chancre), both tests should be repeated after 1 month (1, D). ment.47,50,57,58 These older tests also did not detect cardiolipin However, CLIA and EIA used in many European settings have antibodies (i.e. patients with active infectious syphilis). New suboptimal specificity, resulting in a low positive predictive POCTs have substantially better performances for detection of value in low-prevalence population.22,49,56 If such tests are – both treponemal and non-treponemal antibodies.59 64 Use of used, additionally a reflex confirmatory test by TPHA or rapid POCTs is very important in the WHO strategy for global TPPA should be performed (1, C). elimination of congenital syphilis and mother-to-child-transmis- • In the case a NTT alone is used as a primary screening test, sion (MTCT) of both syphilis and HIV because they permit a positive test must be followed by a reflex TT on the same screening and treatment at the same visit in the field or at periph- serum. If quantitative NTT was not initially done, the NTT eral clinics remote from laboratories (1, D). Currently, where should be repeated quantitatively (1, B). appropriate laboratory diagnostics are available for syphilis in • In the case both a TT and a NTT are used as primary Europe, syphilis POCTs are not recommended for use. Neverthe- screening tests such as (EIA/ELISA/CLIA/TPHA/TPPA plus less, they are useful for on-site testing of outreach populations VDRL/RPR), the NTT must be performed quantitatively (if and in antenatal settings where women with no confirmed syphi- not initially done) in case of positive or discrepant screening lis tests during pregnancy can be tested before delivery. tests (1, B). • The IgG immunoblot for Treponema pallidum has no added Primary screening test(s)4,22,47–49,64–68 major value to other TT. It is expensive and interpretation • TT [TPHA, MHA-TP, TPPA or EIA/ELISA/CLIA] – a TT- of undetermined immunoblot is elusive (1–4 bands). based screening algorithm, using by preference an automa- tized EIA/ELISA/CLIA, is used in many large, well-re- Tests for serological activity of syphilis and for monitoring the sourced European laboratories and is particularly suitable effect of treatment for automated high-throughput screening of asymptomatic • Quantitative VDRL or RPR tests are widely used for moni- populations including blood/plasma donors. The algorithm toring the disease progression and effect of treatment at fol- identifies persons with previous successful treatment of low-up visits. syphilis and those with untreated syphilis. It is usually more • A quantitative titre must be obtained on the very first day sensitive in detecting very early syphilis compared to the use of treatment, that is, to provide a baseline for measuring of a screening NTT. However, it can also result in a high subsequent changes in antibody titres (1, C). number of false-positive tests (i.e. very low positive predic- • Serum should be obtained at 1 month, 3 months and every tive value) in low-prevalence populations. 6 months subsequently, ideally the same NTT should be • NTT [RPR or VDRL] – a NTT-based screening algorithm, used and the samples examined in the same laboratory. This preferably quantitative (i.e. to detect prozone phenomenon should be continued until the NTT becomes negative or in infectious syphilis), is still recommended in some coun- reaches a low plateau (1:1–1:4 sustained for 1 year in the tries. In this algorithm, only active (infectious) syphilis is absence of ongoing risk) (2, C). Patients with persistent detected; however, it has a lower sensitivity compared to high titres should remain under follow-up. using a TT as primary screening test, and in particular, very early syphilis can be missed. Laboratory: false-negative syphilis serology4,22,37,38,43. • TT combined with a NTT – this algorithm is particularly • All STS (TT and NTT) are negative before the appearance useful in cases where the suspicion of very early syphilis is of a chancre and in the first about 5–15 days of the chancre. high (recent chancre, contacts of syphilis cases etc.), because Both TT and NTT can be positive or negative or discor- in some patients NTT may become reactive before TT. dance can occur as follows: positive TT/negative NTT (2/3 of cases in primary syphilis) or negative TT/positive NTT Confirmatory test(s) if any screening test is (1/3 of cases in primary syphilis).43 A negative NTT (or pre- positive4,22,47–49,64–68 sent at a low titre plateau, see above) along with a positive • In the case, a TT being used alone as a primary screening TT is frequently seen in treated and cured syphilis. Of note, test, if positive, a confirmatory TT of a different type is of particularly in late syphilis NTT can remain positive despite limited value in informing treatment,69,70 but a reflex provision of adequate treatment.

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 2020 European Syphilis guideline 5

• A persistent false-negative TT in the course of the disease is positive reactions at low titres of antibodies. Retesting on a exceedingly rare and can usually be explained by technical second sample is necessary in case of negative NTT. problems in the laboratory testing or mix up of samples. • A false-negative NTT (along with positive TT) may occur Laboratory tests to confirm or exclude neurosyphilis15,24,75-87. A especially in early syphilis due to the prozone phenomenon complete clinical examination (neurological, ocular and oto- (an excess of antibodies) when using undiluted serum. logic) is recommended in every patient with positive STS. How- Dilution of serum for NTT must be performed in each case ever, in those without symptoms it is rarely of a positive TT, at least to 1:8 or better 1:16.71 This point contributory.79,81,83,86-88 may be of particular importance if the index/optical density • Fundoscopy must be performed before lumbar puncture units of EIA/ELISA/CLIA is high, and clinicians and labora- (LP). Computer tomography (CT) of the brain should be tory personnel must ensure the NTT titration has been per- requested if neurological problems are identified. formed effectively (1, B). • CSF assessment is not indicated in early syphilis (whether • A false-negative NTT has also been described in old text- the patient is HIV positive or negative), unless there are books in active (very) late-stage syphilis (Bordet–Wasser- neurological, ocular or auricular symptoms (1, A). mann reaction). This is an extraordinarly rare situation and • CSF assessment is indicated in patients with: may not occur with modern tests.72,73 - clinical evidence of neurological, ocular and auricular • Temporarily negative NTT and TT (reactive on subsequent involvement, whatever the stage of the disease79 (1, C) testing) have occasionally been reported in secondary syphi- - tertiary syphilis (cardiovascular, gummatous) (1, D) lis (so-called malignant syphilis). The diagnosis should then • The definition of asymptomatic neurosyphilis is extremely be supported by DFE, T. pallidum PCR, histology and/or difficult and contentious. Most definitions depend on a histochemistry. combination of CSF laboratory tests (protein, cells, CSF TT • Retesting with both TT and NTT is necessary on a second and CSF NTT) but no consensual definition exists. serum, when discordant results are found in an asymp- • Although CSF penicillin levels after injection of benzathine tomatic patient. In the case of chancre (in the absence of penicillin G (BPG) are frequently under the reputed peni- clinically overt vesicles), if DFE is positive or not available, cillin treponemicidal level,89 progression from asymp- treatment should be administered in all cases (syndromic tomatic to symptomatic neurosyphilis is extraordinarily approach) before obtaining laboratory results (T. pallidum rare (even in HIV-positive patients).86 As CSF assessment is PCR, Herpes PCR and STS). This recommendation is an not without its own dangers, LP is not recommended in the important safeguard in many settings where follow-up is vast majority of asymptomatic patients (1, D). not optimal (2, C) • Although robust evidence is lacking, some experts still rec- ommend CSF assessment in asymptomatic patients in the Laboratory: false-positive syphilis serology4,22,37,38,74. following situations for exclusion of asymptomatic neu- • Biological false-positive (BFP) NTT results are associated rosyphilis (2, D): with various medical conditions and have been estimated to - in HIV-positive patients with late syphilis AND CD4 occur in 0.2–0.8% of tests (and even higher in some stud- cells ≤ 350/mm3 AND/OR a serum VDRL/RPR titre ies). They can be divided into acute (<6 months) and >1:3287 chronic (≥6 months). Acute BFP may be seen in postimmu- - in those who have serological failure (leprosy, malignancies, - in those given alternative treatment (e.g. tetracyclines such chronic liver pathology and older age. The majority of BFP as doxycycline) for late syphilis NTT sera show antibody titres of ≤1:4. A positive NTT must • Examination of CSF: must include total protein, number of be retested on a second sample along with a TT. mononuclear cells, a TT (TPHA/MHA-TP/TPPA) and a • Occasional BFP TT tests (FTA-abs test more frequently than NTT (preferably VDRL and otherwise RPR).82 TPHA/MHA-TP/TPPA) may be seen in autoimmune dis- - A normal protein level is possible in neurosyphilis. eases, Lyme disease and possibly during pregnancy. It can - The number of mononuclear cells in CSF can be normal in be excluded with, for example, the IgG immunoblot test for neurosyphilis, especially in parenchymatous neurosyphilis T. pallidum. All TT requiring visual reading of results (tabes dorsalis, general paresis).75,76,87 Conversely, high (FTA-abs test, TPHA, TPPA...) are more subject to false- number of mononuclear cells in CSF can be observed in a

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 6 Janier et al.

number of situations, including HIV infection in the assessed by local epidemiology) are present. Assessment and vac- absence of syphilis. cination for hepatitis B should also be considered if appropriate. - A positive CSF VDRL test is observed in only about 1:3rd of cases of neurosyphilis but a positive test can in the General remarks87,92–97 absence of substantial blood contamination be considered • A treponemicidal level of antimicrobial should be achieved as indicative of neurosyphilis in late disease. However, in in the serum and in the case of neurosyphilis also strived for early syphilis the significance of a positive CSF VDRL test as much as feasible in the CSF. A penicillin level of is less clear.83 >0.018 mg/L is considered treponemicidal, but this level is - A positive CSF TT (TPHA/TPPA) does not confirm the substantially lower than the maximally effective in vitro con- diagnosis of neurosyphilis but a negative CSF TT means centration (0.36 mg/L). neurosyphilis is highly unlikely.15,84,85 • The duration to maintain a treponemicidal level of antimicro- - Several indices can be calculated to evaluate the signifi- bials should be at least 7–10daystocoveranumberofbacterial cance of CSF anti-treponemal immunoglobulins, taking generation times (30–33 h). Longer duration of treatment is into account transfer through the blood–brain barrier. needed as the duration of infection increases (more relapses However, none of these have proven to be of significant have been seen in later stages after short courses of treatment), practical use.4 possibly because of more slowly dividing treponemes in late • PCR assays for detection of T. pallidum in CSF to help syphilis (2, D). Treponemes have been shown to persist despite establish a diagnosis of neurosyphilis are currently consid- apparently successful treatment.93 The significance of this find- ered of limited value since they have shown low sensitivity ing, if any, remains unknown. and suboptimal specificity.15,24,90 • In general, long-acting BPG 2.4 million units is the treat- • In case of an abnormal CSF examination (high protein level ment of first choice, which provides a treponemicidal peni- and/or hypercytosis), repeat CSF control must be per- cillin concentration in blood for up to 21 days. With daily formed after treatment (6 weeks-6 months). parenteral treatment with procaine penicillin, a ‘safety mar- gin’ is provided by giving courses lasting 10–14 days in Investigation for cardiovascular syphilis early syphilis and 10–21 days in late syphilis. However, • Any patient with aortic insufficiency or thoracic aortic well-controlled clinical data are lacking on the optimal dose, aneurysm should be screened for syphilis. duration of treatment and long-term efficacy of all antimi- • Auscultation must be performed in patients with late latent crobials, even for BPG and other penicillins. or tertiary syphilis and in patients with latent syphilis of • Treatment recommendations are based mainly on laboratory unknown duration. A chest X-ray is rarely contributory.91 considerations, biological plausibility, practical considerations, expert opinions, case studies and past clinical experience. Investigation for ocular syphilis • Parenteral rather than oral penicillin treatment is the treatment • Any patient with unexplained sudden visual loss should be of choice because parenteral therapy is supervised with guaran- screened for syphilis. teed bioavailability. However, amoxicillin given orally in combi- • Clinical ocular assessment must be performed in patients with nation with probenecid appears to be effective and results in secondary, early latent, tertiary and late latent syphilis, and a treponemicidal drug levels within the CSF.97,98 fundoscopy performed if any clinical ocular signs are found. • Non-penicillin antibiotics have been evaluated. These • Performing CSF examination is controversial as intravenous include tetracyclines (doxycycline is the preferred tetracy- (IV) penicillin therapy will be initiated anyway. It may be cline with good penetration into the CSF) and macrolides, – helpful to exclude other pathologies in the differential diag- taken orally.99 103 Doxycycline has been evaluated more nosis and if found to be abnormal in someone with neu- than any other non-penicillin antibiotics but all studies have – rosyphilis, appropriate follow-up is required to ensure all been observational and retrospective.100 103 Newer anti-tre- markers return to acceptable levels (2, C). ponemal antibiotics include the intramuscular or intra- venous extended-spectrum cephalosporin (ESC) – Investigation for auricular syphilis ceftriaxone.103 106 Ceftriaxone has good CSF penetration, Any patient with unexplained sudden hearing loss should be but it requires multiple injections, the dose and duration screened for syphilis. are not standardized and it does not offer any advantages over single-dose BPG.107 However, like oral doxycycline, Management daily ceftriaxone injected intravenously or subcutaneously Individuals with syphilis are at higher risk of acquiring other may be an alternative in patients with bleeding disorders. STIs and should have a full STI assessment. All patients with • In case of penicillin allergy, use of ceftriaxone may be a dan- syphilis should also be tested for HIV and HCV if risk factors (as gerous option although cross-allergies are not frequent.

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 2020 European Syphilis guideline 7

History of penicillin anaphylaxis is an absolute contraindica- Replacing part (i.e. 0.5–1 cc) of the diluent by lidocaine 1% tion.65 The oral ESC cefixime is currently evaluated for treat- solution without epinephrine may reduce the discomfort ment, but appropriate data are still pending. associated with injection,118 although this is not feasible in • Azithromycin has shown good treponemicidal activity in animal case of premounted BPG syringes. studies and several controlled studies, mostly in Africa. How- Patients should be kept for 30 min for clinical review after ever, rapid emergence of resistance to azithromycin and clinical injection. – failures have been described in several studies.33,108 113 Although there are more than ten different pharmaceutic • The host immune response is important as 60% of untreated companies manufacturing BPG in Europe, shortages and sup- patients will not develop clinical features other then primary ply disruptions are common.119 lesions.114 CSF involvement is common in early syphilis.76,90 Second-line therapy option Although both parenteral BPG and standard regimens of • Procaine penicillin 600 000 units IM daily for 10–14 days, parenteral procaine penicillin do not achieve treponemicidal i.e. if BPG is not available (1, C) CSF levels,77,84 the prevalence of late syphilis, including neu- rosyphilis, remains low, indicating that treatment at current Bleeding disorders doses is effective and suggesting that host immune responses • Ceftriaxone 1g intravenously (IV) in a single daily dose for in early syphilis play an essential part. 10 days (1, C) • BPG is widely used because of efficacy and ease of treat- • Doxycycline 200 mg daily (either 100 mg twice daily or as a ment. Replacing part of the diluent by the same volume of single 200 mg dose) orally for 14 days (1, C) 1% lidocaine solution may reduce the pain associated with 115 injection and in late syphilis may improve compliance Penicillin allergy or parenteral treatment refused with the second and third injection. Compliance with daily • Doxycycline 200 mg daily (either 100 mg twice daily or as a intramuscular injections with procaine penicillin has been single 200 mg dose) orally for 14 days (1, C) shown to be good in the United Kingdom.116 The control of • Desensitization to penicillin is an option but not possible in syphilis over the past 50 years has been excellent compared many settings and labour intensive. to the prepenicillin era and late complications of syphilis Late latent (i.e. acquired ≥1 year previously or of unknown and/or failures of treatment are uncommon, even in duration), cardiovascular and gummatous syphilis patients with concomitant HIV infection. • There is no established relationship between immune-sup- First-line therapy option pression and the severity of the syphilis-related disease. • BPG 2.4 million units IM, given as one injection of 2.4 mil- However, a closer follow-up (i.e. 1, 3, 6, 9 and 12 months) can lion units or two separate injections of 1.2 million units in be recommended in HIV-positive patients, particularly if the each buttock, on day 1, 8 and 15 (1, C) + ≤ 3 CD4 cell count is 350/mm and/or if the patient is not treated Replacing part (i.e. 0.5–1 cc) of the diluent by lidocaine 1% with antiretroviral therapy (2, D). HIV coinfection does not solution without epinephrine may reduce the discomfort appear to increase the risk of developing a more aggressive course associated with injection,118 although this is not feasible in of early syphilis. Modest differences have been published with a case of premounted BPG syringes. slightly higher prevalence of (1) multiple chancres and (2) con- Patients should be kept for 30 min for clinical review after comitant chancre and secondary eruption in patients infected injection. with HIV. The risk of ocular and neurological involvement is not increased in HIV-positive patients with early syphilis. Thus CSF Second-line therapy option assessment in early syphilis is indicated only in patients with overt • Procaine penicillin 600 000 units IM daily for 17–21 days, ocular, auricular or neurologic symptoms (for the same reason as i.e. if BPG is not available (1, C) in non-HIV-infected patients) (1, A)4,65,66 Penicillin allergy or parenteral treatment refused Some special- Recommended treatment regimens (Table 1) ists recommend penicillin desensitization because the evidence base for the effectiveness of non-penicillin regimens is weak. Early syphilis (Primary, Secondary and Early latent, i.e. • Doxycycline 200 mg daily (either 100 mg twice daily or as a acquired <1 year previously) single 200 mg dose) orally for 21–28 days (2, D)

First-line therapy option2,4,66,90,117 Neurosyphilis, ocular and auricular syphilis • BPG 2.4 million units intramuscularly (IM), given as one • Regimens that achieve treponemicidal levels of an antibiotic injection of 2.4 million units or two separate injections of in the CSF should be the treatment of choice: IV therapy is 1.2 million units in each buttock, on day 1 (1, B) the best option.

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 8 Janier et al.

• Other regimens with weaker evidence can also achieve tre- Prevention of congenital syphilis by serological screening dur- ponemicidal levels in the CSF, i.e. the procaine penicillin/ ing pregnancy and preventive neonatal treatment probenecid combination (IM) and ceftriaxone (IV or IM). • All pregnant women should be screened at their first The availability of probenecid may be a problem. antenatal visit (first trimester) (1, C). Serology should be • Early ocular syphilis such as uveitis syphilitica of short repeated, ideally during third trimester at 28–32 weeks’ duration may be successfully treated with BPG but this gestation and at delivery, in cases where there is option is not recommended. increased risk and in settings with a high syphilis preva- lence (1, C). Furthermore, for pregnant women with no First-line therapy option documented previous test, testing should be performed • Benzyl penicillin 18–24 million units IV daily, as 3–4 mil- at delivery. lion units every 4 h for 10–14 days (1, C) Congenital syphilis Second-line therapy option If hospitalization and IV benzyl fi 120,121,123–125 penicillin is impossible. Con rmed congenital infection • • Ceftriaxone 1–2 g IV in a single daily dose for 10–14 days (1, C) Congenital syphilis is confirmed by identifying T. pallidum • Procaine penicillin 1.2–2.4 million units IM daily AND pro- by DFE or PCR in the placenta or autopsy material, exudate benecid 500 mg four times daily, both for 10–14 days (1, C) from suspicious lesions or body fluids, e.g. nasal discharge.

Penicillin allergy Presumed congenital infection A presumptive diagnosis of • Desensitization to penicillin (in fact, induction of tolerance) congenital syphilis is made in: • followed by the first-line regimen (1, C) A stillborn neonate with a positive TT for syphilis. • Children with a positive TT for syphilis in combination Special considerations with one or more of the following: - persistent rhinitis, condylomata lata, osteitis, periostitis, Pregnancy osteochondritis, ascites, cutaneous and mucous mem- In pregnant women with untreated early syphilis, 70–100% of brane lesions, hepatitis, hepatosplenomegaly, glomeru- infants will be infected, with stillbirths in up to one-third of lonephritis, haemolytic anaemia; – cases.120 122 - radiological abnormalities of the long bones suggestive of Women with persistently negative NTT results are very unli- congenital syphilis; kely to transmit syphilis during pregnancy.123 In case of a posi- - a positive RPR/VDRL test in the cerebrospinal fluid; tive TT along with a negative NTT, repeat NTT after one month - a fourfold increase or more of the TPPA/TPHA titre in to eliminate a very early syphilis. Most transmissions to the fetus the child’s as opposed to the mother’s serum (both occur in late pregnancy (after 28 weeks) and treatment before obtained simultaneously at birth); this period will usually prevent congenital features.120 - a fourfold increase or more of the titre of RPR/VDRL in the child’s as opposed to the mother’s serum (both First-line option for treatment of early syphilis (i.e. acquired obtained simultaneously at birth); <1 year previously) - a fourfold increase or more of the titre of RPR/VDRL in • BPG 2.4 million units IM single dose (or 1.2 million units the child within 3 months after birth; in each buttock) (1, B) - a positive anti-treponemal IgM-EIA, 19S-IgM-FTA-abs Note: some specialists recommend 2 doses of BPG 2.4 million test and/or IgM immunoblot for T. pallidum in the units (days 1 and 8) but the evidence to support this recommen- child’s serum; dation is limited.123 - a mother, in whom syphilis was confirmed during preg- Patients should be observed for adverse reactions to penicillin nancy, but who was not adequately treated either before for 30 min after injection. or during pregnancy. • A child >12 months of age with a positive TT for syphilis Second-line therapy option and in whom sexual abuse has been excluded. • Procaine penicillin 600 000 units IM daily for 10–14 days, i.e. if BPG is not available (1, C). Late congenital syphilis Late congenital syphilis is diagnosed based on: Penicillin allergy • Clinical features: interstitial keratitis, Clutton’s joints, • Desensitization to penicillin followed by the first-line regi- Hutchinson’s incisors, mulberry molars, high palatal arch, men (1, C) perioral rhagades, deafness, frontal bossing, short maxilla,

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 2020 European Syphilis guideline 9

protuberance of mandible, saddle nose deformity, stern- • No data are available concerning the risk of neurosyphilis in oclavicular thickening, paroxysmal cold haemoglobinuria, HIV-infected patients with late syphilis. neurological or gummatous involvement. • Serological tests: these can initially be negative in infants Treatment of syphilis in patients with concomitant HIV infec- infected in late pregnancy and should be repeated. When tion the mother is treated during the last trimester of pregnancy, • Treatment should be given as for non-HIV-infected patients the treatment can be inadequate for the child and the child with careful follow-up to ensure an appropriate response. may still develop congenital syphilis. Syphilis induced by solid organ transplant Investigations First-line therapy options • RPR/VDRL (quantitative), TPPA/TPHA (quantitative), • BPG 2.4 million units IM (one injection 2.4 million units anti-treponemal IgMÀEIA, treponemal IgM (19S-IgM- single dose or 1.2 million units in each buttock) weekly on FTA-abs or IgM immunoblot) – from infant’s blood and days 1, 8 and 15 (1, B)10,135 not umbilical cord blood, since false-positive and false-neg- ative tests may result. Second-line therapy option • Blood: Full blood count, liver function, electrolytes • Procaine penicillin 600 000 units IM daily for 10–14 days, • CSF: cells, protein, RPR/VDRL, TPHA/TPPA i.e. if BPG is not available (1, C)135 • X-rays long bones • Ophthalmic assessment as indicated Penicillin allergy Some specialists recommend penicillin desen- sitization because the evidence base for the effectiveness of non- First-line therapy option penicillin regimens is weak. • Benzyl penicillin 150 000 units/kg IV daily (administered in • Doxycycline 200 mg daily (either 100 mg twice daily or as a – six doses every 4 h) for 10 14 days (1, D) single 200 mg dose) orally during 21–28 days (1, C)135

Second-line therapy option (only if CSF is normal) Reactions to treatment • BPG 50 000 units/kg IM (single dose) up to the adult dose of Patients should be warned of possible reactions to treatment. 2.4 million units (1, D) or procaine penicillin 50 000 units/ Facilities for resuscitation should be available in the treatment kg IM daily for 10–14 days, i.e. if BPG is not available (1, D) area.

HIV-infected patients Jarisch–Herxheimer reaction • An acute febrile illness with headache, myalgia, chills and – General remarks87,90,126 134 rigours, resolving within 24 h. • Serological tests for syphilis in patients with HIV coinfec- • Common in early syphilis (10–25%)136 but is usually not tion are generally reliable for the diagnosis of syphilis and important unless there is neurological or ophthalmic for evaluation of treatment response. involvement, in neonates or in pregnancy when it may • Patients with HIV coinfection may have a slower rate of cause fetal distress and premature labour. decline of VDRL/RPR after treatment but this should not • May be more frequent with penicillin than with doxycy- be considered as failure of response to treatment. cline.136 • False-negative and false-positive tests and delayed • Uncommon in late syphilis but can potentially be life- appearance of seroreactivity have been reported anecdo- threatening if involvement of strategic sites (e.g. coronary tally. ostia, larynx, nervous system).137,138 • In HIV-infected individuals with clinical suspicion of syphi- • Prednisolone can prevent the febrile episode.139 Although lis and negative syphilis serology (confirmed when unproven, biological plausibility suggests that steroids may repeated), it is advisable to perform other diagnostic tests, help to prevent severe deterioration in optic neuritis and e.g. histological, immunofluorescent or PCR examination uveitis following a reaction to treatment in early syphilis. of a biopsy from a clinically suspected lesion or DFE or • Management: PCR to identify treponemes in the exudate from early - If cardiovascular or neurological involvement (including 65 syphilitic lesions. optic neuritis) exists, inpatient management is advisable. • HIV-infected patients with early syphilis do not appear - Prevention of Jarisch–Herxheimer reaction: Prednisolone to have an increased risk of (early) neurological and 20–60 mg daily for 3 days, starting syphilis treatment 24 h ocular involvement or higher rate of treatment failure after commencing prednisolone (2, D) with BPG. • Antipyretics

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 10 Janier et al.

Procaine reaction (procaine psychosis, procaine mania, Hoigne information resources can be found at https://iusti.org/patie syndrome) nt-information/. • Due to inadvertent IV injection of procaine penicillin, the • Although the division of latent syphilis in early and late risk of which may be minimised by the ‘aspiration tech- stages is useful for determining treatment and partner noti- nique’ of injection. fication, this classification can be problematic for use in • Characterized by fear of impending death and may cause surveillance, as a substantial number of late, hypothetically hallucinations or fits immediately after injection. Lasts less non-infectious, latent syphilis cases (latent syphilis of than 20 min. unknown duration classified as late latent) can be due to • Management: early, infectious, latent syphilis. - Exclude anaphylaxis • Sexual contact notification helps to reduce the disease bur- - Calm and verbal reassurance; restraint may be necessary. den in the community, identify asymptomatic syphilitic - Diazepam 5–10 mg rectally/IV/IM if convulsions patients and can delineate relevant sexual risk networks. The use of Internet-based notification systems and commu- Anaphylactic shock nity outreach programmes may assist in identifying and • Facilities for treatment of anaphylaxis should be available as engaging with sexual partners. penicillin is a common cause. • Sexual contacts should include all those individuals who • Management: have had oral, genital or anal intercourse with infected - Epinephrine (adrenaline) 1:1000 (1 mg/mL) IM 0.5 mL individuals, whether or not barrier protection was used. followed by: • For patients with primary syphilis, sexual contacts within ▪ injectable antihistamine, e.g. chlorpheniramine 10 mg IM/ the past 3 months should be notified as the incubation per- IV or clemastine 2–4 mg IM/IV iod is up to 90 days.145 Partner notification may have to be ▪ hydrocortisone 100 mg IM/IV extended to 2 years for patients with secondary syphilis associated with clinical relapse or in early latent syphilis. Pre-exposure prophylaxis and syphilis Longer time periods may be required in those with late Pre-exposure prophylaxis (PrEP) with tenofovir disoproxil latent and tertiary syphilis.144 – fumarate emtricitabine is effective for reducing HIV acquisition • 46–60% of traced sexual contacts, including pregnant – – but may at the individual level result in an increase in the women, of patients with early syphilis are likely to be frequency of condomless sex and several studies have shown it infected.146 140,141 could increase the acquisition of STIs in general. • Immediate epidemiological treatment for sexual contacts This PrEP population should be tested for syphilis every should be considered (especially of pregnant partners!) 3 months. A few studies have evaluated pre-exposure (or imme- unless contacts are able to attend regularly for exclusion of diate post-exposure) prophylactic treatment of STIs using doxy- syphilis through clinical and serological examination (0, 142 143 cycline, taken continuously or intermittently. Although 6 weeks and 3–6 months). this has been shown to potentially decrease the incidence of • Serological tests for syphilis should be performed at the first syphilis and chlamydia, follow-up has been short and there is a visit and repeated at 6 weeks and 3–6 months. risk of promoting development or acquisition of tetracycline • Notification of syphilis to the relevant national authority is resistance in STI bacteria and the microbiome more generally. mandatory in most European countries, particularly early Additional studies in different settings and with longer follow- syphilis and congenital syphilis. The ECDC is responsible up are required before this prophylactic treatment can be recom- for the EU/EEA-wide surveillance of communicable diseases 144 mended. including syphilis.

Contact tracing, management of sexual partners Follow-up and test of cure and notification of syphilis cases The follow-up of treated syphilis patients to ensure cure and • All patients with syphilis should be seen for sexual contact detect reinfection or relapse is achieved by assessing the clinical notification (notification by the patient [patient referral] or and serological response to treatment. Globally, many studies by a health department: provider referral), health education have confirmed that follow-up is suboptimal.126,147 and confirmation of any past treatment history. Further • Early syphilis, minimum clinical and serological (VDRL/ advice from International Union against STI (IUSTI) on RPR) follow-up at 1 month, 3 months then at 6 and this matter can be found in the IUSTI guideline on Partner 12 months. management at https://iusti.org/treatment-guidelines/. - After treatment of early syphilis, the titre of a NTT taken at • Clear information about syphilis should be given to all day 0 (e.g. VDRL and/or RPR) should decline by ≥2 dilution infected individuals and their sexual contacts. Patient steps (≥fourfold decrease in titre of antibodies) within

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 2020 European Syphilis guideline 11

6 months.1,4,22 However, around 15% of patients with early All possible care has been undertaken to ensure the publica- syphilis and no HIV infection do not have a fourfold decrease tion of the correct dosage of medication and route of adminis- – of titre at 6 months, the significance of which is unknown.148 tration. However, it remains the responsibility of the prescribing 150 These patients should be tested again at 12 months. physician to ensure the accuracy and appropriateness of the Patients with repeat early syphilis have usually higher titres medication they prescribe. and a slower decline in titres post-treatment.101,151,152 Acknowledgements - If a fourfold decrease in the antibody titre of a NTT does We are grateful for valuable input on the present guideline to not occur after 6–12 months, (‘serological failure’) some Jonathan Ross, Jorgen Skov Jensen, Agnieszka Serwin and Keith professionals recommend additional treatment with one Radcliffe. weekly injection of BPG 2.4 million units for 3 weeks but Current composition of the European Guideline Editorial no robust evidence for this recommendation exists (2, D) Board can be found at https://iusti.org/treatment-guidelines/ - A negative NTT can be obtained in a substantial number of List of contributing organizations can be reviewed at https:// (but not in all) patients treated for early syphilis after 1– iusti.org/treatment-guidelines/ 2 years. Patients with repeat syphilis serorevert less often. A negative NTT after treatment is considered to be the best Search Strategy and Grading Levels of Evidence confirmation of cure. In patients with persistent low titres Appendix 1. (i.e. ≤4), in NTT (named the serofast state) strict follow-up is recommended but in the absence of ongoing risk these References patients should be considered as successfully treated. In 1 European Centre for Disease Prevention and Control. http://www.ecdc.e patients with persistent high titres of NTT (i.e. ≥8), CSF uropa.eu/. assessment should be considered with the aim of detecting 2 World Health Organization. WHO guidelines for the treatment of Tre- asymptomatic neurosyphilis, although there is no robust ponema pallidum (syphilis) 2016 http://www.who.int/reproductivehea lth/publications/rtis/syphilis-treatment-guidelines/en. evidence for this recommendation (2, D) 3 Spiteri G, Unemo M, Mardh O, Amato-Gauci AJ. The resurgence of - A TT may remain positive for life even following effective syphilis in high-income countries in the 2000s: a focus on Europe. Epi- treatment; and appropriate documentation is necessary to demiol Infect 2019; 147: e143. https://doi.org/10.1017/ prevent unnecessary retreatment. S0950268819000281. • 4 Janier M, Hegyi V, Dupin N, Unemo M, Tiplica GS, French P et al. 2014 In late (latent) syphilis, the serological response of NTTs is European guideline on the management of syphilis. J Eur Acad Dermatol often absent. In non-HIV-infected late latent syphilis Venereol 2014; 20: 300–309. patients with a reactive NTT, which remains stable in a low 5 Grassly NC, Fraser C, Garnett GP. Host immunity and synchronized epi- 433 – titre range, follow-up after treatment is generally not indi- demics of syphilis across the United Stades. Nature 2005; :417 421. 6 Centers for Diseases Control and Prevention. Syphilis (Treponema pal- cated (2, D). lidum) 2018. Case definition. CSTE 17-ID-11. • An increase of ≥2 dilution steps (fourfold increase in anti- 7 Hook EW, 3rd. Syphilis. Lancet 2017; 389: 1550–1557. body titre) in a NTT, in the absence of clinical symptoms, 8 Dupin N. Syphilis. Rev Med Interne 2016; 37: 735–742. suggests reinfection or hypothetic relapse. Treatment should 9 Peeling RW, Mabey D, Kamb ML, Chen XS, Radolf JD, Benzaken AS. Syphilis. Nat Rev Dis Primers 2017; 3: 17073. https://doi.org/10.1038/nrd be given according to the above guidelines for latent syphilis p.2017.73. (early if <1 year; late if ≥1 year) (1, C) Patients at high risk 10 Kovacs CS, Jr, Koval CE, van Duin D et al. Selecting suitable solid organ of reinfection should be checked frequently using NTT, e.g. transplant donors: reducing the risk of donor-transmitted infections. World J Transplant 2014; 4:43–56. every 3 months. (2, C). Reinfection or relapse should be 11 Rompalo AM, Lawlor J, Seaman P, Quinn TC, Zenilman JM, Hook EW. retreated preferably with supervised treatment schedules to Modification of syphilitic genital ulcer manifestations by coexistent HIV ensure compliance and sexual partners should be rescreened. infection. Sex Transm Dis 2001; 28: 448–454. • Follow-up examination of CSF should be performed 12 Hope-Rapp E, Anyfantakis V, FouereSet al. Etiology of genital ulcer disease. A prospective study of 278 cases seen in an STD clinic in Paris. 6 weeks to 6 months after treatment of neurosyphilis to Sex Transm Dis 2010; 37: 153–158. 153 monitor decrease of white blood cells and proteins. (2, 13 Towns JM, Leslie DE, Denham J, Azzato F, Fairley CK, Chen M. Painful D) Limited data suggest that CSF control could be avoided and multiple anogenital lesions are common in men with Treponema in patients with normalization of serum RPR titres.154 pallidum PCR-positive primary syphilis without Herpes simplex virus coinfection: a cross-sectional clinic-based study. Sex Transm Infect 2016; 92: 112–115. Qualifying statement 14 Farhi D, Benhaddou N, Grange P et al. Clinical and serologic baseline The recommendations in this guideline may not be appropriate and follow up features of syphilis according to HIV status in the post 88 – for use in all clinical situations. Decisions to follow these recom- HAART era. Medicine 2009; : 331 340. 15 Vanhaecke C, Grange P, Benhaddou N et al. Clinical and biological mendations must be based on the professional judgement of the characteristics of 40 patients with neurosyphilis and evaluation of Tre- clinician and consideration of individual patient circumstances ponema pallidum nested polymerase chain reaction in cerebrospinal and available resources. fluid samples. Clin Infect Dis 2016; 63: 1180–1188.

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 12 Janier et al.

16 Dombrowski JC, Pedersen R, Marra CM, Kerani R, Golden MR. Preva- and epidemiology of the causative agent of syphilis. Peer J 2019; 6: lence estimates of complicated syphilis. Sex Transm Dis 2015; 42: 702– e6182. 704. 37 Larsen SA, Steiner BM, Rudolph AH. Laboratory diagnosis and interpre- 17 Park KH, Lee MS, Hong IK et al. Bone involvement in secondary syphi- tation of tests for syphilis. Clin Microbiol Rev 1995; 8:1–21. lis: a case report and systematic review of the literature. Sex Transm Dis 38 Nandwani R, Evans DTP. Are you sure it’s syphilis? A review of false 2014; 41: 532–537. positive serology. Int J STD AIDS 1995; 6: 241–248. 18 Mishra S, Walmsley SL, Loutfy MR, Kaul R, Logue KJ, Gold WL. Otosy- 39 Young H. Guidelines for serological testing for syphilis. Sex Transm philis in HIV-coinfected individuals: a case series from Toronto, Infect 2000; 76: 403–405. Canada. AIDS Patient Care STDS 2008; 22: 213–219. 40 Hunter M, Robertson PW, Post JJ. Significance of isolated reactive tre- 19 Davey DJ, Kojima N, Konda KA et al. Transient aortitis documented by ponemal chemiluminescence immunoassay results. J Infect Dis 2013; positron emission tomography in a case series of men and transgender 207: 1416–1423. women infected with syphilis. Sex Transm Infect 2017; 93: 551–555. 41 Cole MJ, Perry KR, Parry JV. Comparative evaluation of 15 serological 20 MMWR: CDC case definitions for public health surveillance. Oct 19 assays for the detection of syphilis infection. Eur J Clin Microbiol Infect 1990/vol.39/No.RR-13. Dis 2007; 26: 705–713. 21 Wheeler HL, Agarwal S, Goh BT. Dark ground microscopy and trepone- 42 Binnicker MJ, Jespersen DJ, Rollins LO. Treponema-specific tests for mal serological tests in the diagnosis of early syphilis. Sex Transm Infect serodiagnosis of syphilis: comparative evaluation of seven assays. J Clin 2004; 80: 411–414. Microbiol 2011; 49: 1313–1317. 22 Ballard R, Hook EW, III. Syphilis. In: Unemo M, Ballard R, Ison C, 43 Gratzer B, Pohl D, Hotton AL et al. Evaluation of diagnostic serological Lewis D, Ndowa F, Peeling R, eds. Laboratory diagnosis of sexually results in cases of suspected primary syphilis infection. Sex Transm Dis transmitted infections, including human immunodeficiency virus. Gen- 2014; 41: 285–289. eva: World Health Organization (WHO). 2013; 107–129. 44 Busse C, Navid MH, Strubel A, Schnitzler P. Evaluation of a new recom- 23 Grange PA, Gressier L, Dion PL et al. Evaluation of a PCR test for detec- binant antigen-based Virotech Treponema pallidum screen ELISA for tion of Treponema pallidum in swabs and blood. J Clin Microbiol 2012; diagnosis of syphilis. Clin Lab 2013; 59: 523–529. 50: 546–552. 45 Marangoni A, Nardini P, Foschi C et al. Evaluation of the BioPlex 2200 24 Marks M, Lawrence D, Kositz C, Mabey D. Diagnostic performance of syphilis system as a first-line method of reverse-sequence screening for PCR assays for the diagnosis of neurosyphilis: a systematic review. Sex syphilis diagnosis. Clin Vaccine Immunol 2013; 20: 1084–1088. Transm Infect 2018; 94: 585–588. 46 Castro A, Jost H, Cox D et al. A comparison of the analytical level of 25 Gayet-Ageron A, Sednaoui P, Lautenschlager S, et al. Use of Treponema agreement of nine treponemal assays for syphilis and possible implica- pallidum PCR in testing of ulcers for diagnosis of primary syphilis. tions for screening algorithms. BMJ Open 2013; 3: e003347. Emerg Infect Dis 2015; 21: 127–129. 47 Sena~ AC, White B, Sparling PF. Novel Treponema pallidum serologic 26 Shields M, Guy RJ, Jeoffreys NJ, Finlayson RJ, Donovan B. A longitudinal tests: a paradigm shift in syphilis screening for the 21st century. Clin evaluation of Treponema pallidum PCR testing in early syphilis. BMC Infect Infect Dis 2010; 51: 700–708. Dis 2012; 12: 353. https://doi.org/10.1186/1471-2334-12-353. 48 Park IU, Chow JM, Bolan G, Stanley M, Shieh J, Schapiro JM. Screening 27 Buffet M, Grange PA, Gerhardt P et al. Diagnosing Treponema pallidum for syphilis with the treponemal immunoassay: analysis of discordant in secondary syphilis by PCR and immunohistochemistry. J Invest Der- serology results and implications for clinical management. J Infect Dis matol 2007; 127: 2345–2350. 2011; 204: 1297–1304. 28 Muller€ H, Eisendle K, Br€auninger W, Kutzner H, Cerroni L, Zelger B. 49 Centers for Disease Control and Prevention (CDC). Discordant results Comparative analysis of immuno-histochemistry, polymerase chain from reverse sequence syphilis screening–five laboratories, United States, reaction and focus-floating microscopy for the detection of Treponema 2006–2010. MMWR Morb Mortal Wkly Rep 2011; 60: 133–137. pallidum in mucocutaneous lesions of primary, secondary and tertiary 50 Park IU, Fakile YF, Chow JM et al. Performance of treponemal tests for syphilis. Br J Dermatol 2011; 165:50–60. the diagnosis of syphilis. Clin Infect Dis 2019; 68: 913–918. 29 Petrich A, Rojas P, Schulze J et al. Fluorescence in situ hybridization for 51 Smajs D, Strouhal M, Knauf S. Genetics of human and animal uncul- the identification of Treponema pallidum in tissue sections. Int J Med tivable treponemal pathogens. Infect Genet Evol 2018; 61:92–107. Microbiol 2015; 305: 709–718. 52 Gray RR, Mulligan CJ, Molin BJ et al. Molecular evolution of the tprC, 30 Grange PA, Allix-Beguec C, Chanal J et al. Molecular subtyping of Tre- D, I, K, G and J genes in the pathogenic genus Treponema. Mol Biol Evol ponema pallidum in Paris, France. Sex Transm Dis 2013; 40: 641–644. 2006; 23: 2220–2233. 31 Peng RR, Wang AL, Li J, Tucker JD, Yin YP, Chen XS. Molecular typing 53 Centurion-Lara A, Giacani L, Godornes C et al. Fine analysis of genetic of Treponema pallidum: a systematic review and meta-analysis. PLoS diversity of the tpr gene family among treponemal species, subspecies Negl Trop Dis 2011; 5: e1273. and strains. PLoS Negl Trop Dis 2013; 7: e2222. 32 Ho EL, Lukehart SA. Syphilis: using modern approaches to understand 54 Tsuboi M, Nishijima T, Aoki T et al. Usefulness of automated latex an old disease. J Clin Invest 2011; 121: 4584–4592. turbidimetric rapid plasma reagin test for diagnosis and evaluation of 33 Popisilova P, Grange PA, Grillova L et al. Multi-locus sequence typing treatment response in syphilis in comparison with manual card test: a of Treponema pallidum subsp. pallidum present in clinical samples from prospective cohort study. J Clin Microbiol 2018; 56: e01003–e1018. France: infecting treponemes are genetically diverse and belong to 18 55 Osbak K, Abdellati S, Tsoumanis A et al. Evaluation of an automated allelic profiles. PLoS One 2018; 13: e0201068. https://doi.org/10.1371/ quantitative latex immunoturbidimetric non-treponemal assay for diag- journal.pone.0201068. nosis and follow-up of syphilis: a prospective cohort study. J Med Micro- 34 Beale MA, Marks M, Sahi SK et al. Genomic epidemiology of syphilis biol 2017; 66: 1130–1139. reveals independent emergence of macrolide resistance across multiple 56 Malm K, Andersson S, Fredlund H et al. Analytic evaluation of nine circulating lineages. Nat Commun 2019; 10: 3255–3263. serological assays for diagnosis of syphilis. J Eur Acad Dermatol Venereol 35 Vrbova E, Grillov L, Mikalova L et al. MLST typing of Treponema pal- 2015; 29: 2369–2376. lidum subsp. pallidum in the Czech Republic during 2004–2017: Clinical 57 Herring AJ, Ballard RC, Pope V et al. A multi-centre evaluation of nine isolates belonged to 25 allelic profiles and harboured 8 novel allelic vari- rapid, point-of-care syphilis tests using archived sera. Sex Transm Infect ants. PLoS One 2019; 14: e0217611. 2006; 82(Suppl 5): v7–12. 36 Grillova L, Jolley K, Smajs D, Picardeau M. A public data base for the 58 Gliddon HD, Peeling R, Kamb ML, Toskin I, Wi TE, Taylor MM. A sys- new MLST scheme for Treponema pallidum subsp. pallidum: surveillance tematic review and meta-analysis of studies evaluating the performance

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 2020 European Syphilis guideline 13

and operational characteristics of dual point-of-care tests for HIV and 80 Marra CM, Tantalo LC, Maxwell CL, Ho EL, Sahi SK, Jones T. The rapid syphilis. Sex Transm Infect 2017; 93(S4): S3–S15. plasma reagin test cannot replace the venereal disease research labora- 59 Owusu-Edusei K, Jr, Gift TL, Ballard RC. Cost-effectiveness of a dual tory test for neurosyphilis diagnosis. Sex Transm Dis 2012; 39: 453–457. non-treponemal/treponemal syphilis point-of-care test to prevent 81 Tuddenham S, Ghanem KG. Neurosyphilis: knowledge gaps and contro- adverse pregnancy outcome in sub-Saharan Africa. Sex Transm Dis versies. Sex Transm Dis 2018; 45: 147–151. 2011; 38: 997–1003. 82 Marra CM, Maxwell CL, Smith SL et al. Cerebrospinal fluid abnormali- 60 Castro AR, Esfandiari J, Kumar S et al. Novel point-of-care test for ties in patients with syphilis: association with clinical and laboratory fea- simultaneous detection of nontreponemal and treponemal antibodies in tures. J Infect Dis 2004; 189: 369–376. patients with syphilis. J Clin Microbiol 2010; 48: 4615–4619. 83 Shi M, Peng RR, Gao Z et al. Risk profiles of neurosyphilis in HIV-nega- 61 Yin YP, Chen XS, Wei WH et al. A dual point-of-care test shows good tive patients with primary, secondary and latent syphilis: implications performance in simultaneously detecting nontreponemal and trepone- for clinical intervention. J Eur Acad Dermatol Venereol 2016; 30: 659– mal antibodies in patients with syphilis: a multisite evaluation study in 666. China. Clin Infect Dis 2013; 56: 659–665. 84 Libois A, De Wit S, Poll B et al. HIV and syphilis: when to perform a 62 Causer L, Kaldor JM, Conway DP et al. An evaluation of a novel dual lumbar puncture. Sex Transm Dis 2007; 34: 141–144. point-of care test for syphilis as a tool to distinguish active from past 85 Harding AS, Ghanem KG. The performance of cerebrospinal fluid tre- treated infection. Clin Infect Dis 2015; 61: 184–191. ponemal-specific antibody tests in neurosyphilis: a systematic review. 63 Owusu-Edusei K, Jr, Koski KA, Ballard RC. The tale of two serologic Sex Transm Dis 2012; 39: 291–297. tests to screen for syphilis – treponemal and non-treponemal: does the 86 Tomkins A, Ahmad S, Cousins D, Thng CM, Vilar FJ, Higgins SP. order matter? Sex Transm Dis 2011; 38: 448–456. Screening for asymptomatic neurosyphilis in HIV: an observational 64 Owusu-Edusei K, Jr, Peterman TA, Ballard RC. Serologic testing for study. Sex Transm Infect 2018; 94: 337–339. syphilis in the United States: a cost-effectiveness analysis of two screen- 87 Ghanem KG. Management of adult syphilis: Key questions to inform the ing algorithms. Sex Transm Dis 2011; 38:1–7. 2015 Centers for Disease Control and Prevention Sexually Transmitted 65 Stoner B. Current controversies in the management of adult syphilis. Diseases treatment guidelines. Clin Infect Dis 2015; 61: S819–836. Clin Infect Dis 2007; 44(Suppl 3): S130–146. 88 Dabis R, Radcliffe K. What is the role of a full physical examination in 66 Workowski KA, Bolan GA. Centers for Disease Control and Prevention the management of asymptomatic patients with late syphilis. Int J STD (CDC). Sexually transmitted diseases treatment guidelines, 2015. AIDS 2012; 23: 901–902. MMWR Recomm Rep 2015; 64:34–51. 89 Mohr JA, Griffiths W, Jackson R et al. Neurosyphilis and penicillin levels 67 Peng J, Lu Y, Yu H et al. Analysis of 2 reverse syphilis testing algorithms in cerebrospinal fluid. JAMA 1976; 236: 2208–2209. in diagnosis of syphilis: a large-cohort prospective study. Clin Infect Dis 90 Rolfs RT, Joesoef MR, Hendershot EF et al. A randomized trial of 2018; 67: 947–953. enhanced therapy for early syphilis in patients with and without 68 Tong ML, Lin LR, Liu LL et al. Analysis of 3 algorithms for syphilis sero- human immunodeficiency virus infection. N Engl J Med 1997; 337: diagnosis and implications for clinical management. Clin Infect Dis 307–314. 2014; 58: 1116–1124. 91 Dabis R, Radcliffe K. Is it useful to perform a chest X-ray in asymp- 69 Cantor AG, Pappas M, Daeges M, Nelson HD. Screening for syphilis. tomatic patients with late syphilis. Int J STD AIDS 2011; 22: 105–106. Updated evidence report and systematic review for the US Preventive 92 Rolfs RT. Treatment of syphilis, 1993. Clin Infect Dis 1995; 20(Suppl 1): Services Task Force. JAMA 2016; 315: 2328–2337. S23–38. 70 Berry GJ, Loeffelholz MJ. Use of treponemal screening assay strength of 93 Dunlop EMC. Survival of treponemes after treatment, comments, clini- signal to avoid unnecessary confirmatory testing. Sex Transm Dis 2016; cal conclusions and recommendations. Genitourin Med 1985; 61: 293– 43: 737–740. 301. 71 Liu LL, Lin LR, Tong ML et al. Incidence and risk factors for the pro- 94 Lowhagen€ GB, Brorson J-E, Kaijser B. Penicillin concentrations in cere- zone phenomenon in serologic testing for syphilis in a large cohort. Clin brospinal fluid and serum after intramuscular, intravenous and oral Infect Dis 2014; 59: 384–389. administration to syphilitic patients. Acta Derm Venereol 1983; 63:53– 72 Wohrl€ S, Geusau A. Neurosyphilis is unlikely in patients with late latent 57. syphilis and a negative blood VDRL test. Acta Derm Venereol 2006; 86: 95 Goh BT, Smith GW, Samarasinghe L, Singh V, Lim KS. Penicillin con- 335–339. centrations in serum and cerebrospinal fluid after intramuscular injec- 73 Tuddenham S, Obeng C, Ghanem KG. Neurosyphilis and ophthalmic tion of aqueous procaine penicillin 0.6 MU with and without oral syphilis in persons with negative rapid plasma reagin and positive tre- probenecid. Br J Venereol Dis 1984; 60: 371–373. ponemal antibody test results. Sex Transm Dis 2015; 42: 347–349. 96 Schoth PEM, Wolters EC. Penicillin concentrations in serum and CSF 74 Geusau A, Kittler H, Hein U, Dangl-Erlach E, Stingl G, Tschachler E. during high-dose intravenous treatment for neurosyphilis. Neurology Biological false-positive tests comprise a high proportion of Venereal 1987; 37: 1214–1216. Disease Research Laboratory reactions in an analysis of 300 000 sera. Int 97 Faber WR, Bos JD, Tietra PJGM, Fass H, van Ejk RTW. Treponemicidal J STD AIDS 2005; 16: 722–726. level of amoxycillin in cerebrospinal fluid after oral administration. Sex 75 Hooshmand H, Escobar MR, Kopf SW. Neurosyphilis; a study of 241 Transm Dis 1983; 10: 148–150. patients. JAMA 1972; 219: 726–729. 98 Tanizaki R, Nishijima T, Aoki T et al. High-dose oral amoxicillin plus 76 Lowhagen€ GB, Andersson M, Blomstrand C, Roupe G. Central nervous probenecid is highly effective for syphilis in patients with HIV infection. system involvement in early syphilis. Part I. Intrathecal immunoglobulin Clin Infect Dis 2015; 61: 177–183. production. Acta Derm Venereol 1983; 63: 409–417. 99 Ghanem KG, Erbelding EJ, Cheng WW, Rompalo AM. Doxycycline 77 Wiesel J, Rose DN, Silver AL, Sacks HS, Bernstein RH. Lumbar puncture compared with benzathine penicillin for the treatment of syphilis. Clin in asymptomatic late syphilis. Arch Intern Med 1985; 145: 465–468. Infect Dis 2006; 42: e45–49. 78 Lukehart SA, Hook EW, III, Bakerzander SA, Collier AC, Critchlow 100 Antonio MB, Cuba GT, Vasconcelos RP et al. Natural experiment of CW, Handsfield HH. Invasion of the central nervous system by Tre- syphilis treatment with doxycycline or benzathine penicillin in HIV-in- ponema pallidum: implications for diagnosis and treatment. Ann Intern fected patients. AIDS 2019; 33:77–81. Med 1988; 109: 855–862. 101 Salado-Rasmussen K, Hoffmann S, Cowan S et al. Serological response 79 Davis AP, Stern J, Tantalo L et al. How well do neurologic symptoms iden- to treatment of syphilis with doxycycline compared with penicillin in tify individuals with neurosyphilis. Clin Infect Dis 2018; 66: 363–367. HIV-infected individuals. Acta Derm Venereol 2016; 96: 807–811.

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 14 Janier et al.

102 Tsai JC, Lin HY, Lu PL et al. Comparison of serological response to 122 Hong FC, Wu XB, Yang F et al. Risk of congenital syphilis (CS) follow- doxycycline versus benzathine-penicillin G in the treatment of early ing treatment of maternal syphilis: results of a CS control program in syphilis in HIV-infected patients: a multicentre observational study. China. Clin Infect Dis 2017; 65: 588–594. PLoS One 2014; 9: e109813. 123 Peterman TA, Newman DR, Davis D, Su JR. Do women with persis- 103 Liu H-Y, Han Y, Chen X-S et al. Comparison of efficacy of treatments tently negative nontreponemal test results transmit syphilis during preg- for early syphilis: a systematic review and network meta-analysis of ran- nancy? Sex Transm Dis 2013; 40: 311–315. domized controlled trials and observational studies. PLoS One 2017; 12: 124 Walker GJA. Antibiotics for syphilis diagnosed during pregnancy. e0180001. Cochrane Database Syst Rev 2001; 3: CD001143. https://doi.org/10.1002/ 104 Marra CM, Boutin P, McArthur JC et al. A pilot study evaluating ceftri- 14651858.CD001143. axone and penicillin G as treatment agents for neurosyphilis in human 125 Herremans T, Kortbeek L, Notermans DW. A review of diagnostic tests immunodeficiency virus-infected individuals. Clin Infect Dis 2000; 30: for congenital syphilis in newborns. Eur J Clin Microbiol Infect Dis 2010; 540–544. 29: 495–501. 105 Dowell ME, Ross PG, Musher DM, Cate TR, Baughn RE. Response of 126 Blank LJ, Rompalo AM, Erbelding EJ, Zenilman JM, Ghanem KG. latent syphilis or neurosyphilis to ceftriaxone therapy in persons infected Treatment of syphilis in HIV-infected subjects: a systematic review of with human immunodeficiency virus. Am J Med 1992; 93: 481–488. the literature. Sex Transm Infect 2011; 87:9–16. 106 Cao Y, Su X, Wang Q et al. A multicentre study evaluating ceftriaxone 127 Janier M, Chastang C, Spindler E et al. A prospective study of the influ- and benzathine penicillin G as treatment agents for early syphilis in ence of HIV status on the seroreversion of serological tests for syphilis. Jiangsu. China. Clin Infect Dis 2017; 65: 1683–1688. Dermatology 1999; 198: 362–369. 107 Smith NH, Musher DM, Huang DB et al. Response of HIV infected 128 Sena~ AC, Wolff M, Martin DH et al. Predictors of serological cure and patients with asymptomatic syphilis to intensive intramuscular therapy serofast state after treatment in HIV-negative persons with early syphilis. with ceftriaxone or procaine penicillin. Int J STD AIDS 2004; 15: 328– Clin Infect Dis 2011; 53: 1092–1099. 332. 129 Ghanem KG, Erbelding EJ, Wiener ZS, Rompalo AM. Serological 108 Riedner G, Rusizoka M, Todd J et al. Single dose azithromycin versus response to syphilis treatment in HIV-positive and HIV-negative penicillin G benzathine for the treatment of early syphilis. N Engl J Med patients attending sexually transmitted disease clinics. Sex Transm Infect 2005; 353: 1236–1244. 2007; 83:97–101. 109 Hook EW, Behets F, Van Damme K et al. A phase III equivalence trial of 130 Frohlich-Knaute€ D, Graf N, Lautenschlager S, Weber R, Bosshard PP. azithromycin versus benzathine penicillin for treatment of early syphilis. Serological response to treatment of syphilis according to disease stage J Infect Dis 2010; 201: 1729–1735. and HIV status. Clin Infect Dis 2012; 55: 1615–1622. 110 Lukehart SA, Godornes C, Molini BJ et al. Macrolide resistance in Tre- 131 Gonzalez-Lopez JJ, Fernandez-Guerrero ML, Lujan R, Fernandez-Tos- ponema pallidum in the United States and Ireland. N Engl J Med 2004; tado S, de Gorgolas M, Requena L. Factors determining serologic 351: 154–158. response to treatment in patients with syphilis. Clin Infect Dis 2009; 49: 111 Stamm LV. Global challenge of antibiotic-resistant Treponema pallidum. 1505–1511. Antimicrob Agents Chemother 2010; 54: 583–589. 132 Courjon J, Hubiche T, Dupin N, Grange PA, Del Giudice P. Clinical 112 Zhou P, Li K, Lu H et al. Azithromycin treatment failure among primary aspects of syphilis reinfection in HIV-infected patients. Dermatology and secondary syphilis patients in Shanghai. Sex Transm Dis 2010; 37: 2015; 230: 302–307. 726–729. 133 Ganesan A, Mesner O, Okulicz JF et al. A single dose of benzathine peni- 113 Chen XS, Yin YP, Wei WH et al. High prevalence of azithromycin resis- cillin G is as effective as multiple doses of benzathine penicillin G for the tance to Treponema pallidum in geographically different areas in China. treatment of HIV-infected persons with early syphilis. Clin Infect Dis Clin Microbiol Infect 2013; 19: 975–979. 2015; 60: 653–660. 114 Gjestland T. The Oslo study of untreated syphilis; an epidemiologic 134 Costa-Silva M, Azevedo C, Azevedo F, Lisboa C. Early syphilis treatment investigation of the natural course of the syphilitic infection based upon in HIV-infected patients: single dose vs. three doses of benzathine peni- a re-study of the Boeck-Bruusgaard material. Acta Derm Venereol Suppl cillin G. J Eur Acad Dermatol Venereol 2016; 10: 1805–1809. (Stockh) 1955; 35(Suppl 34):3–368; Annex I-LVI. 135 Cortes NJ, Afzali B, MacLean D et al. Transmission of syphilis by solid 115 Amir J, Ginat S, Cohen YH, Marcus TE, Keller N, Varsano I. Lidocaine organ transplantation. Am J Transplant 2006; 6: 2497–2499. as a diluent for administration of benzathine penicillin G. Pediatr Infect 136 Arando M, Fernandez-Naval C, Mota-Foix M et al. The Jarisch-Herx- Dis 1998; 17: 890–893. heimer reaction in syphilis: could molecular typing help to understand 116 Crowe G, Theodore C, Forster GE, Goh BT. Acceptability and compli- it better? J Eur Acad Dermatol Venereol 2018; 32: 1791–1795. ance with daily injections of procaine penicillin in the treatment of 137 Degos R, Lortat-Jacob E, Macrez P. Aorto-myocardite syphilitique sero- syphilis-treponemal infection. Sex Transm Dis 1997; 24: 127–130. positive. Bull Soc Fr Dermatol Syphil 1970; 77:32–33. 117 Wong T, Singh AE, De P. Primary syphilis treatment response to doxy- 138 Belum GR, Belum VR, Arudra SKC et al. The Jarisch-Herxheimer reac- cycline/tetracycline versus benzathine penicillin. Am J Med 2008; 121: tion: revisited. Travel Med Infect Dis 2013; 11: 231–237. 903–908. 139 Gudjonsson H, Skog E. The effect of prednisolone on the Jarisch-Herx- 118 Janier M, Libar E, Bonnet A et al. Treatment of late syphilis with 2.4 mil- heimer reaction. Acta Dermatol Venereol 1968; 48:15–18. lion units benzathine penicillin G (BPG): tolerance of single versus 140 Beymer M, DeVost MA, Weiss RE et al. Does HIV pre-exposure prophy- divided doses. Sex Transm Dis 2012; 39: 359–360. laxis use lead to a higher incidence of sexually transmitted infections? A 119 Nurse-Findlay S, Taylor MM, Savage M et al. Shortages of benzathine case-crossover study of men who have sex with men in Los Angeles, Cal- penicillin for prevention of mother-to-child transmission of syphilis: an ifornia. Sex Transm Infect 2018; 94: 457–462. evaluation from multi-country surveys and stakeholder interviews. PLoS 141 Traeger MW, Schroeder SE, Wright EJ et al. Effects of pre-exposure pro- Med. 2017; 14: e1002473. phylaxis for the prevention of Human Immunodeficiency Virus infec- 120 Boot JM, Oranje AP, De Groot R, Tan G, Stolz E. Congenital syphilis. tion on sexual risk behaviour in men who have sex with men: a Int J STD AIDS 1992; 3: 161–167. systematic review and meta-analysis. Clin Infect Dis 2018; 67: 676–686. 121 Kamb ML, Newman LM, Riley PL, Mark J, Hawkes SJ, Malik T et al.A 142 Bolan RK, Beymer MR, Weiss RE, Flynn RP, Leibowitz AA, Klausner JD. road map for the global elimination of congenital syphilis. Obstet Doxycycline prophylaxis to reduce incident syphilis among HIV-infected Gynecol Int 2010; 2010: 312798. https://doi.org/10.1155/2010/ men who have sex with men who continue to engage in high-risk sex: a 312798. randomized, controlled pilot study. Sex Transm Dis 2015; 42:98–103.

JEADV 2020 © 2020 European Academy of Dermatology and Venereology 2020 European Syphilis guideline 15

143 Molina JM, Charreau I, Chidiac C et al. Post-exposure prophylaxis with A Grade 1 recommendation is a strong recommendation doxycycline to prevent sexually transmitted infections in men who have to do (or not do) something, where benefits clearly out- sex with men: an open-label randomised substudy of the ANRS IPER- GAY trial. Lancet Infect Dis 2018; 18: 308–317. weigh risks (or vice versa) for most, if not all, patients. 144 Grant JS, Stafylis C, Celum C et al. Doxycycline prophylaxis for bacterial Most clinicians and patients would want to follow a strong sexually transmitted infections. Clin Infect Dis 2020; 70: 1247–1253. recommendation unless there is a clear rationale for an 145 Tiplica GS, Radcliffe K, Evans C et al. European Guidelines for the man- alternative approach. A strong recommendation usually starts agement of partners of persons with sexually transmitted infections. J ... Eur Acad Dermatol Venereol 2015; 29: 1251–1257. with the standard wording: ‘We recommend ’ or ‘It is 146 Schober PC, Gabriel G, White P, Felton WF, Thin RN. How infectious is recommended ...’ syphilis? Br J Vener Dis 1983; 59: 217–219. A Grade 2 recommendation is a weaker or conditional recom- 147 Katz KA, Lee MA, Gray T, Marcus JL, Pierce EF. Repeat syphilis among mendation, where the risks and benefits are more closely bal- men who have sex with men-San Diego county, 2004–2009. Sex Transm Dis 2011; 38: 349–352. anced or are more uncertain. Alternative approaches or 148 Sena~ AC, Zang X-H, Li T et al. A systematic review of syphilis serological strategies may be reasonable depending on the individual treatment outcomes in HIV-infected and HIV-uninfected persons: patient’s circumstances, preferences and values. A weak or con- rethinking the significance of serological non-responsiveness and the ditional recommendation usually starts with the standard word- serofast state after therapy. BMC Infect Dis 2015; 15: 479. https://doi. org/10.1186/s12879-015-1209-0. ing: ‘We suggest...’ or ‘It is suggested...’ The strength of a 149 Sena~ AC, Wolff M, Behets F et al. Rate of decline in Nontreponemal recommendation is determined not only by the quality of evi- antibody titers and seroreversion after treatment of early syphilis. Sex dence for defined outcomes but also the balance between desir- Transm Dis 2017; 44:6–10. 150 Pastuszczak M, Jakiela B, Wojas-Pelc A. Association of interleukin-10 able and undesirable effects of a treatment or intervention, promoter polymorphisms with serofast state after syphilis treatment. Sex differences in values and preferences, and, where appropriate, Transm Infect 2019; 95: 163–168. resource use. Each recommendation concerns a defined target 151 Knaute DF, Graf N, Lautenschlager S, Weber R, Bosshard PP. Serologi- population and is actionable. cal response to treatment of syphilis according to disease stage and HIV status. Clin Infect Dis 2012; 55: 1615–1622. The quality of evidence is graded from A to D and is defined 152 Kenyon C, Tsoumanis A, Osbak K et al. Repeat syphilis has a different as follows: immune response compared with initial syphilis: an analysis of biomar- • Grade A evidence means high-quality evidence that comes 94 – ker kinetics in two cohorts. Sex Transm Infect 2018; : 180 186. from consistent results from well-performed randomized 153 Marra CM, Maxwell CL, Tantalo L et al. Normalization of cerebrospinal fluid abnormalities after neurosyphilis therapy: does HIV status matter. controlled trials (RCTs) or overwhelming evidence from Clin Infect Dis 2004; 38: 1001–1006. another source (such as well-executed observational studies 154 Marra CM, Maxwell CL, Tantalo SC, Sahi SK, Lukehart SA. Normaliza- with consistent strong effects and exclusion of all potential tion of serum rapid plasma reagin titer predicts normalization of cere- sources of bias). Grade A implies confidence that the true brospinal fluid and clinical abnormalities after treatment of neurosyphilis. Clin Infect Dis 2008; 47: 893–899. effect lies close to the estimate of the effect. • Grade B evidence means moderate-quality evidence from Appendix 1 randomized trials that suffers from serious flaws in conduct, inconsistency, indirectness, imprecise estimates, reporting Search strategy bias or some combination of these limitations, or from This guideline has been updated from the IUSTI-Europe Syphilis other study designs with specific strengths such as observa- guideline 2014.4 Evidence for this guideline was provided by tional studies with consistent effects and exclusion of the review of the Medline/PubMed, Embase and Cochrane Library majority of the potential sources of bias. from 2014 to October 2019, using the term syphilis, neu- • Grade C evidence is low-quality evidence from controlled rosyphilis, congenital syphilis and Treponema pallidum. trials with several serious limitations or observational stud- ies with limited evidence on effects and exclusion of most The levels of evidence and grading of recommendations potential sources of bias. • Grade D evidence is based only on case studies, expert Levels of evidence judgement or observational studies with inconsistent effects All key recommendations made for diagnosis and management and a potential for substantial bias, such that there can be have been graded for the level of evidence. little confidence in the effect estimate.

JEADV 2020 © 2020 European Academy of Dermatology and Venereology